JPH05244407A - Image processing method - Google Patents

Abstract

PURPOSE: To compress a full color range by storing multi-pixel images provided with the color of a first full color range in a memory and converting the color of the multi-pixel images not present in a second full color range to the color of the second full color range. CONSTITUTION: A full color range comparison device 37 decides a compression amount to source images and stores it in a color mapping buffer 41. A color mapping device 34 reads the source images from an image buffer 33 and maps the color of the source images corresponding to the compression amount stored in the buffer 41. Especially, as soon as the source images are read from the buffer 33 to an image output device 35, the device 34 maps the source images by using a full color range compression amount stored in a color mapping table. As a result, the color of the source images is put inside the full color range of the device 35. When the device 35 receives the output of the device 34, the device 35 outputs the images corresponding to the images for which the full color range is compressed.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION This invention relates to performing full color range compression in devices such as color printer devices. In particular,
The present invention relates to a color printer device in which a color image to be printed is stored and analyzed at the same time, and the entire color gamut of the image is removed from memory for printing and simultaneously compressed.

[0002]

Color printers are known which can be used to produce a color representation of an original image and a color image. Methods and apparatus relating to color image reproduction are described, for example, in U.S. Pat. Nos. 4,670,780 and 4,707,72.
7, No. 4,731,662, No. 4,873,570
It is described in each publication of the issue. For literature on color image reproduction, see Stone et al., "Color Gamut Mapping a.
nd the Printing of Digital Color Images. ", ACM Tra
nsactions on Graphics, Vol. 7, No. 4, pp. 249-292
(October 1988); Stone et al., "Color Gamut Mapping a
nd the Printing of Digital Color Images ", Zerox Re
port EDL-88-1 [p 88-00021], pp. 1-52 (April 1988); G
ordon et al., "On the Rendition of Unprintable Col
ors ", Proceedings of the Technical Association of
There is the Graphic Arts, pp. 186-195 (1987).

Color printers are limited in the range of colors they can reproduce. For example, the source image (ie, the original image) to be printed by a color printer may contain colors that the printer cannot reproduce. Especially CR
Since the total color gamut available for T (cathode ray tube) is typically much larger than that of a color printer,
This is true for color images produced for CRT monitors. Therefore, before printing is performed, the colors of the entire source image must often be mapped to the colors of the entire printer (ie, the colors that the printer can produce). This process involves compressing the full color gamut of the source image until the source image is within the full color gamut of the printer, so it is called "full gamut compression (gam
ut compression) ".

A method for mapping the colors of a source image to colors that can be produced by a printer over a gamut of colors (ie, a source image gamut compression method) is described in US Pat.
It is disclosed in Japanese Patent No. 8,885. This patent publication ('
885), the full color range of the source image on the monitor is contracted, and as a result, the full color range of the printer is included.

However, the above-mentioned patent publication ('88
In 5), the entire color gamut of each source image is shrunk, i.e. compressed, in the same way. Even though the source image contains only colors that the printer can print, the colors in the source image full color gamut are still mapped (ie, the source image full color gamut is still compressed) before the source image is reproduced. ). As described above, in the patent publication ('885), unnecessary distortion and change are introduced into the source image when the original source image entire color range does not require compression because it falls within the printer entire color range. Will end up.

Rather than compressing each source image in the same manner, it is preferable to determine a different mapping or compression scheme for each source image so that unwanted distortion of the source image can be kept to a minimum. For example, if the color of the source image gamut is the printer gamut, then the source image should not be mapped (ie, the gamut of the source image should not be compressed). This approach is taught in Stone et al, supra. However, the Stone et al reference requires a sophisticated human analysis of the source image gamut and the printer gamut.

The sophisticated human analysis described above is described in the literature P. Lai.
hanen "Optimization of Digital Color Reproduction
on the Basis of Visual Assessment of Reproduced Im
ages ", Proceedings of the SID, Vol. 30, No. 3, pp.
183-190 (1989), "Colour Reproduction Theory Based
on the Principles of Color Science ", Advancesin
Printing Science and Technology, June 1987 Confere
nce, Pantech Press, London (1988) can be eliminated by automatically performing full color gamut compression. Although the above-mentioned Laihanen document is intended for automatic compression of the entire color range, it does not mention any technique for performing full-color range compression in real time. The aforementioned US patent publication ('885) also has the same drawbacks. In other words, the colors described in US Pat. No. '885 are not intended to be used to perform real-time adjustment of the full color gamut of various source images or their compression.

Adjusting the full color gamut of the source image to match the full color gamut of the printer is generally a two step process. A conventional device for performing such full color gamut adjustment or compression is shown in FIG. Referring to FIG. 1, the image buffer 13 is generally an image input device 1.
Receives a source image (eg a color image) from 1.
The image is stored as long as it is processed by the full color gamut compression circuit 17. The image input device 11 can be an input for a color monitor, and the image buffer 13 can be a RAM (random access memory) having a storage location corresponding to the pixel of the source image input by the image input device 11. You can At each memory location of the pixel buffer 13, color information corresponding to a given pixel of the source image is stored.

In FIG. 1, the entire color range compression circuit 17 is provided.
Executes a two-step full color range compression process. The first step of the full color gamut compression process involves examining the color of each pixel of the source image stored in the image buffer 13 for the purpose of determining the boundaries of the full color gamut of the source image. The second step, performed after the full color gamut of the source image is compared with the full color gamut of the printer, is to convert the color values of each pixel stored in the image buffer 13 and store it in the image buffer 13. Matching the colors of the source image to the full color gamut of the printer.

In the full color gamut compression process performed by the system shown in FIG.
The color of each pixel stored in must be considered twice. That is, the first time is to analyze the boundary of the entire color range of the source image, and the second time is to map or convert the color of each pixel of the source image to compress the entire color range of the source image. After the image has thus been mapped to the colors of the source image stored in buffer 13 (ie, after the full color range of the source image has been compressed to fall within the full color range of the printer), the image is
To image output device 15 (eg color printer)
Is output to.

In FIG. 2, the color of the image is changed to 2 as in FIG.
2 shows a conventional color printer system that is considered. However, in the system of FIG. 2, full color gamut compression is not performed because source image colors that cannot be printed by the printer system are not input to this system. According to this type of printer system, an operator (eg, a human operator) who uses the printer system does not input a color that the image output device 25 cannot print.
Source image colors that cannot be printed will not be input to the printer system of FIG.

Color images with sophisticated image representations can be described using the conventional page description language described in the following references: In other words, S.Harrington
et al., "Interpress: The Source Book", Simon & Sch
uster, New York (1988); "Interpress Electronic Pri
nting Standard ", XNSS 048601, Xerox Adobe Systems,
Inc., Mass. (1990). These languages can be used in a system such as the printer system shown in FIG.

In the system shown in FIG. 2, a color image is constructed in the image buffer 23 by a plurality of single image units (also called pixels) such as lines and / or characters. The construction of the source image in the image buffer 23 is performed by the image unit generator 21. The image unit generator 21 assembles or builds a color image by collecting various image units in an image buffer 23. When the construction of the color image in the image buffer 23 is completed, the contents of the image buffer 23 are sent to the image output device 25, and as a result, the constructed color image can be printed. The color of the source image is considered twice. That is, the first time is when the image unit enters the image buffer 23, and the second time is when the contents of the image buffer are printed.

The printer system of FIG. 2 is limited to the case where an operator using the image unit generator 21 generates an image unit having a color that the image output device 25 can produce for storage in the image buffer 23. in this way,
In the printer system of FIG. 2, the full color range of the source image stored in the image buffer 13 is always within the full color range of the printer (that is, the full color range of the image output device 25). do not need. Further, since the printer system of FIG. 2 uses only input image units having colors within the entire color range of the image output device 25, the storage capacity of the image buffer 23 is such that the image buffer 23 is within the entire color range of the printer. It is designed to represent or store only included colors.

[0015]

The disadvantage of the conventional printer system shown in FIG. 1 is that the entire color gamut of the source image cannot be determined in real time (ie the source image is being input to the image buffer 13 at the same time), or It cannot be compressed in real time (that is, the source image is being read from the image buffer 13 at the same time). Instead, the total color gamut of the source image is determined by the full color gamut compression circuit 17 after the source image is stored in the image buffer 13. Similarly, the source image is read from the image buffer 13 after compression of the entire color range of the source image.

The printer system of FIG. 2 also has drawbacks.
In the printer system of FIG. 2, the color selection of the source image available to the operator while the operator is constructing the source image is limited by the characteristics of the image output device 25 and the image buffer 23. In other words, the source image constructed in the image buffer 23 by the operator consists only of colors that can be rendered in the image buffer 23 and reproduced by the image output device 25.

[0017]

The present invention has been made in view of the above, and according to one aspect of the present invention, when a source image is read from a memory to an image output device such as a printer. A configuration is provided that can simultaneously perform full color gamut compression on the source image. According to another aspect of the present invention, the compression of the entire color gamut of the source image is simply and efficiently corrected by using the total color gamut compression factor indexed by the hue-related value indicating the hue characteristic of the source image. Provides a configuration performed by. Further, compression of the source image full color gamut is accomplished by representing the values of the source image in the YES friction system, and also using the logarithmic value indicative of the saturation characteristics of the source image and the full color gamut compression factor applied to the source image. Can be simplified by

Other features and advantages of the invention are disclosed and suggested in the following description or are taught by the practice of the invention. The features and advantages of the invention will be realized and attained by means of the constructions and combinations particularly pointed out in the appended claims. As embodied and broadly described herein, the image processing method of the present invention performed by a data processing system having a memory should be output by an image output device capable of outputting a second full color gamut of colors. Storing a multi-pixel image having a first full color gamut in a memory of the data processing system, and reading the multi-pixel image from the memory of the data processing system for reception by the image output device. Converting the color of the multi-pixel image that is not in the second full color gamut to a color of the second full color gamut after the beginning of the reading step and before completion of the read step. To have. Another aspect of the image processing method of the present invention executed by a data processing system having a memory is to be output by an image output device capable of outputting colors in the second full color range.
Storing a multi-pixel image having a first full color gamut in a memory of the data processing system;
A second step of constructing a color mapping table in the image memory, the table including a list of all-color-range compression factors indicating differences in all-color ranges, and storing the color-mapping table in the image memory according to the all-color-range compression factors in the color-mapping table. Modifying the multi-pixel image being processed.

In addition to the above aspects, the present invention also comprises apparatus for performing the above method.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT FIG. 3 illustrates an apparatus for producing a printed color image according to one embodiment of the present invention. A flow chart illustrating the full color gamut compression process performed by the apparatus of FIG. 3 is shown in FIG. The full color gamut compression process performed by the apparatus of FIG. 3 is a two step process. The first step involves analyzing the colors of the source image (eg, determining the boundaries of the entire color gamut of the source image), and the second step maps or transforms the colors of the source image to convert all colors of the source image. Concerning compressing range.

The apparatus of FIG. 3 performs full color gamut compression using two color considerations involved in source image construction and printing. In particular, the first color consideration associated with the construction of the source image (ie, inspection of the source image that can be stored in memory) is used to determine the total color gamut of the source image and the second color consideration associated with printing the source image ( That is, the source image is compressed using a printable stored source image inspection).

In the apparatus of FIG. 3, the digital processing system 100 includes a main processor 30 (eg CP
U), the main processor 30 cooperates with a color mapping device 234, which is preferably an auxiliary processor or external hardware chip of this main processor. In other embodiments of the invention, as described below, the functions of the color mapping device 34 may be performed by the main processor 30 itself rather than an auxiliary processor or separate chip of the main processor 30. The main processor 30 and the color mapping device 34 work together to compress the entire color gamut of the source image and output it via the image output device 35. The image output device 35 is a display device such as a printer or a CRT (cathode ray tube).

As can be seen from FIG. 3, the main processor 30 stores the image buffer 33 for storing the source image and the boundary of the entire color range of the source image (that is, the information indicating the boundary of the entire color range of the source image). It cooperates with the source image full color gamut buffer 36, the image output device full color gamut buffer 40, and the color mapping buffer 41. Buffers 33, 36, 40, 41 may be portions of a single random access memory (RAM) included in digital processing system 100, or, in other embodiments, separate RAMs. In yet another embodiment, the full color range buffer 40 is RA
A read only memory (ROM) may be used instead of M.

The main processor 30 includes an image unit generator 31, a color inspection device 32, and a full color gamut comparison device 37. Although the image unit generator 31, the color inspection device 32, and the full color gamut comparison device 37 are represented as separate elements within the main processor 30 of FIG. 3, these elements are preferably memory (eg, RA.
M, not shown), embodied by software stored in M, which is executed by the main processor 30. The configuration of FIG. 3 can also be embodied by an analog or digital circuit that performs the same function as the software described below.

The image unit generator 31 is the main processor 3
A color source image is formed in the image buffer 33 of 0. The image unit generator 31 forms a source image from a plurality of image units such as lines and characters under the control of an operator (for example, a human operator). Each image unit consists of one or more pixels, and the image unit generator 31 outputs one image unit at a time.

The arrangement implemented by the image unit generator 31, for example, forms the source image from "scratch" or lines and / or images into a given color image output by a scanner or other image generating device. Add. In other embodiments of the present invention, the image unit generator may be eliminated and the signal output by the image unit generator may be output by the scanner or the output of another external image input device of the main processor 30. Can be replaced.

The color inspection device 32 inspects the color of the source image output by the image unit generator 31. In particular, after the image unit generator 31 outputs the image unit to be stored in the image buffer 33, the color inspection device 3
2 examines the color of the image unit before it is received in the image buffer 33. Therefore, while constructing the source image in the image buffer 33, the image unit generator 31
Each image unit output by is inspected by the color inspection device 32 before being stored in the image buffer 33.

As can be seen from the above description, the color inspection device 32 stores the last image unit of the source image in the image buffer 33 from the time when the first image unit of the source image is output from the image unit generator 31. Inspect all image units of the source image until time. Thus, at the same time that the source image generated by the image unit generator 31 is stored in the image buffer 31, the color inspection device 32 inspects or considers the colors of the source image (i.e., the first image of the source image). After the unit has been output for receipt by the image buffer 33, and before the last image unit of the source image has been stored in the image buffer 33).

The color inspection performed by the color inspection device 32 will be described below. As can be seen from FIG. 3, the color inspection device 32 includes two color inspection units 32a and 32b for analyzing the colors of the source image.
In particular, the color inspection unit 32b analyzes the image unit output by the image unit generator 31 to correct which pixel position in the image buffer 33 to determine the color information of the source image. This color information is also used to perform this modification (ie the color coordinates are stored at the pixel location to be modified). The correction of the storage location in the image buffer 33 is executed by the color inspection unit 32b.

The color inspection unit 32a analyzes the source image to determine the boundary of the entire color range of the source image, and also determines the boundary of the entire color range of the source image (that is, information about the boundary of the entire color range of the source image. ) Is stored in the source image full color range buffer 36. In this way, each color inspection unit 32a, 32b inspects or analyzes the color of the source image, while the source image is constructed or stored in the image buffer 33. Further, the source image is stored in the image buffer 33 by the color inspection unit 32b, and at the same time, the color inspection unit 32a stores the boundary of the entire color range of the source image in the source image full color range buffer 36.

In the apparatus of FIG. 3, the color inspection units 32a and 32b of the color inspection device 32 are included in the main processor 30, and therefore the functions performed by these units 32a and 32b are executed in the main processor 30. It However, in another embodiment of the invention, the function performed by one of these sections 32a, 32b is performed by a separate auxiliary processor from the main processor 30 and the color inspection section 32a of the color inspection device 32 is executed. , 32b can also be operated in parallel. In this case, the color of each image unit of the source image can be inspected at the same time by both the color inspection units 32a and 32b of the color inspection device 32. Furthermore, the color inspection unit 3
2b stores the color coordinates for each image unit of the source image in the image buffer 33 in parallel with the total color range information derived from the image unit in the image total color range buffer 36 (executed by the color inspection unit 32a). Is stored. With such parallel processing,
The storage speed of the source image in the image buffer 33 and the storage speed of the boundary of the entire color range of the source image in the source image full color range buffer 36 are increased.

In another embodiment of the present invention, all the functions performed by the color checker 32a can be performed in parallel using a first set of auxiliary processors external to the main processor 30, and All functions performed by the tester 32b can be performed in parallel using a second set of auxiliary processors external to the main processor 30. According to this embodiment, for example, the first set of processors can analyze one of the image units of the source image and, in parallel, output the color space coordinates of the previously analyzed image unit. Using the auxiliary set of processors arranged in a pipeline configuration, the color inspection units 32a, 3
By performing the function of 2b, the storage speed of the source image in the image buffer 33 and the storage speed of the boundary of the entire color range of the source image in the source image full color range buffer 36 are further increased.

The image buffer 33 is addressed according to the pixel location of the source image so that each memory location of the image buffer 33 can maintain color space coordinates for a given one of the source image pixels. The source image full color gamut buffer 36 also includes a color space array formed by memory locations addressed by colors such that each memory location in the source image full color gamut buffer 36 contains a color of the full color gamut of the source image. The information corresponding to a given one of The source image may be transferred to the image buffer 3 by using another address configuration.
1 or the source image full color range can be stored in the source image full color range buffer 36.

The full color range comparison device 37 and the color mapping buffer 41 will be described below. Similar to main processor 30, color mapping device 34 preferably includes elements embodied by software stored in memory (not shown) whose instructions are executed by color mapping device 34. Also,
As mentioned above, the color mapping device 34 can also be embodied by analog circuits that perform the same functions as the software described below.

The full color gamut comparison device 37 cooperates with the source image full color gamut buffer 36, the image output device full color gamut buffer 40, and the color mapping device 41, and the color mapping device 34 includes the color mapping buffer 41 and the color mapping buffer 41. It cooperates with the image buffer 33. The image output device full color range buffer 40 is used to store the entire color range boundaries of the image output device 35 (that is, the image output device full color range boundaries). Preferably,
The image output device full color gamut buffer 40 is addressed similarly to the source image full color gamut buffer 36,
As a result, the full color gamut comparison device 37 can easily compare the full color gamut of the source image with the full color gamut of the image output device. If the image output device 35 is replaced with a different image output device, and the image output device full color range buffer 40 is composed of a RAM, the boundary of the image output device full color range stored in the image output device full color range buffer 40 is imaged. A new image output device corresponding to the image output device replaced by the output device 35 can be replaced with the entire color range boundary.

The full color gamut comparison device 37 compares the boundaries of the source image full color range stored in the source image full color range buffer 36 with the image output device full color range stored in the image output device full color range buffer 40. , As a result, for each color of the source image that does not meet the full color gamut of the image output device, determine a value that indicates how much the color of the non-printable source image must be mapped or compressed to fit the full color gamut of the printer. To do. Such a value will be referred to as “compression amount” or “squeeze (squ
eeze) amount ”.

The full color range comparison device 37 determines the compression amount for the source image and stores the compression amount in the color mapping buffer 41. The color mapping device 34 reads the source image from the image buffer 33 and maps the color of the source image according to the compression amount stored in the color mapping buffer 41. In particular, the source image is the image buffer 3
3 is read out to the image output device 35, and at the same time, the color mapping device 34 maps the source image using the entire color range compression amount stored in the color mapping table (that is, the first pixel of the source image is the image. After being read from the image buffer for reception by the output device and before the last pixel of the source image is actually received by the image output device 35), so that the colors of the source image are all colors of the image output device. It will be within the range. The output of the color mapping device 34 (ie,
When the image output device 35 receives the full-color-range compressed source image), the image-output device 35 outputs an image corresponding to the full-color-range compressed image. The mapped source image supplied to the image output device 35 has a full gamut of compressed colors so that each color of the image can be reproduced by the image output device 35.

The flow chart of FIG. 4 illustrates the full color gamut compression method performed by the apparatus of FIG. The steps of FIG. 4 are performed by the components of FIG. 3 described above. Summarizing the above-described full color range compression processing, the source image is stored in the image buffer 33 to determine the source image full color range, and at the same time, the color inspection device 32 inspects the color of the source image generated by the image unit generator 31. (Step S1 in FIG. 4). After that, the full color gamut comparison device 37 uses the source image full color gamut buffer 3
The boundary of the source image full color range stored in 6 is compared with the boundary of the image output full color range stored in the image output device full color range buffer 40 (step S2). Next, the full color range comparison device 37 determines the full color range compression amount based on the comparison between the boundary of the source image full color range and the boundary of the image output full color range (step S3). Next, the source image is the image buffer 33.
At the same time, the color mapping device 34 maps the colors of the source image (step S4). Similarly, the color inspection units 32 a, 3 of the color inspection device 32
2b, all functions of full color gamut comparison device 37, and color mapping device 34 can be implemented using a set of processors having a pipeline configuration external to main processor 30. Thus, for example, the full color gamut comparison device 37 can perform these functions in parallel (comparison of full color gamut boundaries, output of full color gamut compression amount to the color mapping buffer 41, etc.), and color mapping. The device 34 can execute these functions (color mapping, output of the mapped color to the image output device 35, etc.) in parallel.

An advantage of the apparatus of FIG. 3 is that it allows real-time source image gamut determination and real-time source image compression. The source image is constructed in the image buffer 33, and at the same time the entire color range of the source image is inspected (that is, analyzed), and the source image is read from the image buffer 33 to the image output device 35 and at the same time the source image is mapped. Thus, both the time required to perform source image full color gamut compression and the number of steps required to perform source image full color gamut compression are reduced. In addition, the apparatus shown in FIG. 3 allows greater operator design flexibility by increasing the number of colors used by the operator to build the source image in the image buffer 33. Further, the conventional system compresses the entire color gamut of the source image regardless of whether compression is required, whereas the apparatus of FIG. 3 compresses only the entire color gamut of the source image that requires compression. It is advantageous in terms.

In the apparatus of FIG. 3, the source image is not compressed until it has been stored in the image buffer 33, so the image buffer 33 is "wide" so that it can store most of the possible source image colors. Thus, the above-described embodiment differs from the prior art system of FIG. 2 in which the image buffer 23 represents only colors within the full color gamut of the image output device. Extending the image buffer 33 to include almost all of the possible source image colors either increases the storage capacity of the image buffer 33 to accommodate more bits at each pixel location, or less color levels to the image output device. It means assigning to the entire color gamut and losing some color accuracy. However, the step between color space coordinates outside the color range of the image output device, that is, the step between the color space coordinates outside the color range of the image output device, is calculated using a non-linear scale indicating the color space coordinates stored in the image buffer 33. That is, the loss of color accuracy can be reduced by making the distance larger than the interval.

As mentioned above, the image unit generator 31 is used by a human operator to construct a color source image in the image buffer 33. The color of the image unit output by the image unit generator 31 can be represented by the well-known RGB color coordinate system or CIEXYZ color coordinate system. Also, other well known color coordinate systems can be used to represent the various colors of the source image.

First, color RGB values or CIE
When the XYZ values are converted to the YES coordinate system, analysis of the colors output by the image unit generator 31 and full color range compression are more easily performed. Such transformations occur in the image unit generator 31, the color inspection device 32, or an apparatus (not shown) located between the image unit generator 31 and the path to the color inspection device 32. The color of the image unit to be output by the image unit generator 31 is originally represented in the YES coordinate system, and
No conversion to the YES coordinate system needs to be performed.

The YES coordinate system is based on Xerox Corporation (US Conne
ticut, Stamford) produced a color standard (ie “Xerox Color Encoding Standard”, XNSS288811,
1989). Y representing the color of the source image
The ES coordinate values can be determined from the RGB coordinates of the source image according to the formula below. Y = 0.253R + 0.684G + 0.063B (1) E = (R-G) / 2 (2) S = (R + G) / 4 + B / 2 (3) By the image unit generator 31 The output color coordinate is C
If it is an IE XYZ value, the following conversion formula is applied.

Y = Y (4) E = 2.019X-1.473Y-0.246Z (5) S = 0.423X + 0.277Y-0.831Z (6) In the apparatus of FIG. Y of the output of the generator 31
After conversion to ES coordinates, the color inspection unit 32b of the color inspection device 32 inspects the source image and stores it in the image buffer 33 in the YES coordinate format. At the same time, the color inspection unit 32a of the color inspection device 32 inspects the color of the source image and stores the boundary of the source image full color range in the source image full color range buffer 36 in the YES coordinate format.

FIG. 5 shows an example of the source image full color range 50 and the printer full color range 55 expressed in the YES coordinate system. In the YES coordinate system, Y represents luminescence, E is the red-green axis, and S is the blue-yellow axis. Figure 5
As can be seen, the boundaries of the source image gamut 50 match the YES color coordinates contained in the source image constructed by the image unit generator 31, and the printer gamut 55 boundaries can be reproduced by the image output device 35. Match the color. In FIG. 5, point A indicates the color of the source image within the printer's full color gamut and point B indicates the color of the source image to be compressed or mapped to fall within the printer's full color gamut.

In the preferred embodiment of the invention, FIG.
The source image full color gamut can be stored in the source image full color gamut buffer 36 as a full color gamut representation addressed by color, as shown in FIG. However,
Another efficient way of storing the source image full color gamut in the source image full color gamut buffer 36 (and the image output device full color gamut in the image output device full color gamut buffer 40) is the source image full color gamut. The boundary of the source image gamut and the boundary of the image output device gamut in each of the gamut buffer 36 and the image output device gamut buffer 40 as a set of values associated with saturation in the index tables. To represent.

The other method described above is the color inspection device 32.
Executed by When the color inspecting unit 32a receives the source image color from the image unit generator 31, the color inspecting unit 32a causes the YES coordinate representing the color and the logarithmic value of the absolute value of the E and S coordinate values for the color (the base is, for example, 10). To decide. This process is performed for all the colors of the source image and the resulting values are used to determine the amount of compression used to perform full color range compression in the YES coordinate system using cylindrical geometry. The color inspection device 32 also determines the logarithmic value of the Y-coordinate (for example, the base is 10) (or the logarithmic value of the absolute value of the Y-coordinate) for each color, and thereby the distribution of the source image colors is visually uniform. Can be

The logarithmic value of the absolute value of the E coordinate of each color and the logarithmic value of the absolute value of the S coordinate of each color are determined by the color inspection device 32 as follows. 1e = sign (E) (A log (max (| E |, N)) + B) (7) 1s = sign (S) (A log (max (| S |, N)) + B) (8) le = Logarithmic value of absolute value of E coordinate value ls = Logarithmic value of absolute value of S coordinate value Sign (E) = Sign of E coordinate value (+ or −) Sign (S) = Sign of S coordinate value (+ or −) Max (| E |, N) = | E |, N, whichever is greater max (| S |, N) = | S |, N, whichever is greater A, B, N are E, S value regions And le, a constant representing the range of ls values. Constants A and B are scalings and offsets to extend the le, ls range (eg, 60 and 2.1, respectively). The constant N is a non-zero cutoff value (for example, 0.007812) for avoiding a decision error in a state where the E and S coordinate values pass through 0 (that is, equal to 0).
5). The constants A, B and N can be determined based on the input / output range of the equations (7) and (8). For example,
When the range of the value of the input E (or the input S) is -1 to +1 and the desired range of the value of the output le (or the output Is) is -128 to +128, the constants A, B, and C are used.
Are 60, 2.1 and 0.0078125, respectively. Preferably, the color inspection device 32 uses a look-up table to convert from E to le and S to ls.
Conversion to.

FIG. 6 is a diagram of the source image full color gamut points A, B as viewed along the Y axis of the YES coordinate system shown in FIG. After determining the le, ls values for the color coordinates of each source image color, the color inspection device 32 uses the le, ls values for each quadrant I, II, III, IV of FIG. Establish an index table in.
In FIG. 6, point A of the source image full color range 50 of FIG. 5 is included in quadrant I, and point B of the source image full color range 50 of FIG. 5 is included in quadrant III.

The four quadrants are defined by four possible sign combinations of (sign (E), sign (S). For example, the source image color has a negative sign (E) value and a positive sign (S) value. (That is, E of the source image color
When the value is negative and the S value is positive), its color coordinates are in quadrant II. Since four possible code combinations of the code (E) and the code (S) represent each quadrant in FIG. 6,
The color checking device 32 generates 4 by generating an index table for each of the 4 possible code combinations.
An index table can be generated for one quadrant.

Table 1 is the top of the index table corresponding to the quadrant.

[0052]

[Table 1] The index table for each quadrant is the absolute value of the le coordinate value of the source image color occurring in that quadrant (ie | le
|) List or table, each | le | value being indexed or pointed to by a corresponding H, Y index. The index table in each quadrant stores only the maximum | le | value corresponding to each H, Y index in that quadrant, rather than storing all the | le | values corresponding to each H, Y index. For ease of explanation, Table 1 shows only the top of the index table (ie, a typical index table has many 4 or more | le.
| Can contain values).

Next, the relationship among the H, Y and | le | values of each color will be described with reference to FIG. FIG. 7 shows the point A in the entire color range 50 of the source image in FIG. 5 by the YES coordinate system using cylindrical geometry. As mentioned above, point A is in quadrant I of FIG. Therefore, the | le | value calculated for point A will be stored in the index table set for quadrant I.

Referring to FIG. 7, for the color having the point A, the measured amount that the color is saturated with the | le | value as the color coordinate value of the point A (the horizontal line from the Y axis to the color coordinate of the color) (With respect to length).
In particular, the | le | value as the coordinate value of the color having the point A increases when the length of the horizontal line from the Y axis to the point A increases. The hue of the color can be measured using the H value as the color coordinate of the point A for the color having the point A (that is, the H value represents the hue in relation to the angle of the horizontal line with respect to the Y axis). The Y value as the color coordinate of the point A with respect to the color having the point A is the luminescence or gray level of the color (that is, the height of the color coordinate of the color above the origin of the Y axis of the YES coordinate system).

Returning to Table 1, the | le | values stored in the look-up table for each quadrant are indexed by the H, Y values determined by the color checking device 32. In another embodiment of the present invention, the | le | value can be indexed only by the H value. Furthermore, in another embodiment of the invention, the | le | value can be indexed by a function of the H and Y values, eg the logarithmic value of the Y value.

As mentioned above, the length of the horizontal line from the Y axis to the color coordinate of a given color indicates the amount of color saturation. The color mapping of the source image color, performed to compress the entire color gamut of the source image, shortens the lines associated with a color (that is, shrinks the saturation of the color) to bring the color coordinates of the color into the full gamut of the image output device. It is to put in. | Le | for a given color
Coordinates and | ls | coordinates can be used to measure the line length associated with a color (how the color is saturated), thus reducing the | le | and | ls | values associated with a given color This shortens the line for that color.

The H, Y index for a given quadrant is determined in the color inspection device 32 by determining the H, Y values for each color of the source image in that quadrant. As mentioned above, the Y value for a given color is simply the gray level or Y coordinate of that color.
Thus, a function of the Y value such as the Y value or its logarithmic value can be used as the Y value of the H, Y index. The H value of the H and Y indices for each color is the color inspection device 3
It is decided by 2 as follows.

H = | le | − | ls | (9) where H relates to one hue of the source image color, and |
le | = absolute value of le coordinate associated with the color, | ls |
= The absolute value of the ls coordinate associated with the color. The H value of each color in a given quadrant is a line extending horizontally from the color coordinates of a pixel to the Y axis and the boundary between the quadrant and the S and E axes.
It is equal to the logarithm of the tangent of two. For example, in FIG. 6, the H value associated with point A is one color coordinate of the source image color and is equal to the tangent logarithm of the angle θ _A (eg, base 10). By determining the H, Y values for each color of the source image and the | le | value for each color in each quadrant, the color inspection device 32 comprises a portion similar to the look-up table portion shown in Table 1 for each quadrant. Build an index table in the source image full color gamut buffer 36. The index table for each quadrant is H, Y
| Le of the colors in that quadrant indexed by value
| Is a list of values. In this embodiment, the H and Y values are quantized to provide a table size that is easy to process.

In the preferred embodiment, color inspection device 32 stores only the maximum | le | value for each H, Y index in the index table for each quadrant of source image full color range buffer 36. Therefore, each H, Y value stored in the source image full color gamut buffer 36 touches only a single | le | value in the index table, and the | le | specified by each H, Y index corresponds to that H, Y value. The maximum | le | value that has been recorded for the index. The amount of memory required to store the index table is reduced because each index table stores only the maximum | le | value, rather than all of the | le | values corresponding to each H, Y index. Maximum | le
The | value represents the boundary of the entire color range of the source image.

The maximum | le | value for each H, Y index is stored by the color inspecting device 32 at the next "if-th".
en "statement. if table [H, Y] <| le |, then table [H, Y] = | le | where table [H, Y] = value H, stored in one of the index tables, The current | le | value indexed by Y, | le | = potential displacement of table [H, Y].

Following this "if-then" statement, color checking device 32 compares each new | le | value for a given H, Y index with the current | le | value (ie, table [H, Y]). As a result, when the new | le | value is larger than the current | le | value, [H, Y] in the source image full color range buffer 36.
The current | le | value indexed by is the new | l
e | Replaces the value. In this way, the color inspection device 32
Sets each index table and sets each H that points to the index table,
Only include the largest | le | value that occurs for the Y index.

In another embodiment, the source image full color gamut buffer 36 has a maximum of | l for each H, Y index.
s | Store the value. Further, in another embodiment, the source image full color gamut buffer 36 is | le | for each H, Y index as the value of the index table for each quadrant.
The larger of the value and | ls | value is stored. Further, in another embodiment, the source image gamut buffer 36
Is the maximum | l for the H, Y index in the given quadrant
Store the e | value and store the maximum | ls | value for the remaining H, Y indexes in that quadrant.

Generally, it is desirable to store information indicating maximum saturation corresponding to each H, Y index in the index table for each quadrant. Since the saturation of a given color corresponds to the length of the line extending from the Y axis to the color coordinate of the color, the information stored in the index table for each quadrant identifies the longest line for each H, Y index in that quadrant. is doing. In other words, each index table shows the maximum saturation that occurs for each H, Y index that points to the index table.

The conversion to Y, le, ls coordinates facilitates the determination of four look-up tables representing the source image full color gamut. Together, the four index tables make up one source image full color gamut table stored in the source image full color gamut buffer 36 and represent the boundaries of the source image full color gamut. Full color range comparison device 37
Compares the boundary of the source image full color range with the boundary of the image output device full color range using the source image full color range table stored in the source image full color range buffer 36. In particular, the full color gamut comparison device 37 uses the source image full color gamut table stored in the source image full color gamut buffer 36 to determine the source image represented by the | le | value using the maximum of the source image full color gamut. The boundary of the entire color gamut is compared with the boundary of the image output device full color range represented by the maximum | le | value of the image output device full color range, thereby mapping the color of the source image stored in the image buffer 33. Determine the appropriate amount of compression to use. Image output device The boundary of the full color gamut is described by a table that is substantially the same as the source image full color gamut table (eg, by a table consisting of multiple index tables each containing a saturation indexed value in both H and Y values). , The difference between the source image full color gamut table value and the image output device full color range table and values should be such that the source image full color range falls within the image output device full color range. Indicates the degree. In particular,
The difference between the source image full color range table value and the image output device full color range table value is the difference between the source image saturation value (eg, the source image | le | value) and the value reproducible by the image output device 35. Generates logarithmic value.

The difference F [H, Y] between the value from the source image full color range table and the value from the image output device full color range table is determined by the full color range comparison device 37 by the following equation. F [H, Y] = max (0, table [H, Y] -printer table [H, Y]) (10) where table [H, Y] = index table of the source image full color range table The value H in one,
Current | le | value indexed by value Y, printer table [H, Y] = current | l indexed by value H, value Y in image output device full color range table
e | value, max (0, table [H, Y] -printer table [H, Y]) = 0 and table [H, Y] -printer table [H, Y], whichever is larger F [H, Y] = | Le corresponding to the table [H, Y] value
| Value and corresponding to printer table [H, Y] value | le
Is the logarithmic value of the difference from the value (for example, the base is 10).

After the full color gamut device 37 has determined F [H, Y] for each H, Y index in the source image full color gamut table, the color mapping device 34 is F [H, Y].
The Y] value is used to map (ie, compress) the source image stored in image buffer 33 so that all colors of the source image fall within the full color gamut of the image output device. This will be described later. F [H,
The Y] value is stored in the color mapping device 41 by the full color gamut comparison device 37, so that the source image is read from the image buffer 33 to the image output device 35 while the color of the source image is transferred to the color mapping device 3.
4 (i.e. after reading the first pixel of the source image from the image buffer and before receiving the last pixel by the image output device).

The F [H, Y] values are stored in the color mapping buffer 41 in the color mapping table format by the full color range comparison device 37. F [H, Y] stored in the color mapping table of the color mapping buffer 41, like the saturation value stored in the index table portion shown in Table 1.
The values are indexed by the corresponding H and Y values, respectively.
F [H, Y] indexed by H, Y index
A typical color mapping table containing the values is shown in Table 2.

[0068]

[Table 2] The smoothing operation is applied to the color mapping table stored in the color mapping buffer 41, and as a result, the F [H, Y] value stored in the color mapping table is H,
A large change does not occur with a change from the Y index to the adjacent H and Y indexes. For example, only one of the F [H, Y] values for a given H, Y index in the color mapping table is small (eg, close to 0) and F [X, X for other H, Y indexes close to it. Y] When the value is large,
It is preferable to increase the small F [X, Y] value of the color mapping table to approach the magnitude of the other F [X, Y] values described above.

The details of the method for actually compressing the entire color range of the source image by the color mapping device 34 using the cylindrical geometry will be described below. As mentioned above, the compression or color mapping of the source image gamut performed by the color mapping device 34 is performed by the printer gamut until the color coordinates of the source image gamut are all within the gamut of the image output device. | Le |, | ls | with each source image color having outer color coordinates
It is to reduce the value.

To compress the source image color using cylindrical geometry (ie, | le |, |
ls | value) to reduce the color mapping device 3
4 determines the H, Y index for the source image color, and uses the H, Y index to perform the lookup function of the color mapping table stored in the color mapping buffer 41 to obtain the source image to be compressed. Each color ｜
le |, | ls | value to which F of the color mapping table
Select whether to subtract the [X, Y] value. The color mapping device 34 uses each | le |, | ls of the source image to be compressed.
| Subtracts the selected F [H, Y] value from the value and desaturates the source image color as described below.
To

In order to desaturate each source image color associated with a given H, Y index to the same extent (ie, reduce saturation), the color mapping table 34 stores each source image color associated with the H, Y index. Subtract the | le |, | ls | value with the F [H, Y] value selected for its H, Y index. In other words, the color mapping device 34 subtracts the same F [X, Y] value from the | le | and | ls | values of each source image color specified by the given H, Y index. The color mapping device 34 has the following formula (11),
Perform these subtractions according to (12).

| Le ′ | = max (0, | le | −F [X, Y]) (11) where max (0, | le | −F [X, Y]) = 0,
| Le | -F [X, Y], whichever is larger, | le | = | le | value of the color from the compressed image buffer 33,
F [X, Y] = | le | the amount of compression in the color mapping table with the same H, Y index, | le ′ | = | le of the color after full color range compression has been performed by the color mapping device 34. | Value.

| Ls ′ | = max (0, | ls | −F [X, Y]) (12) where max (0, | ls | −F [X, Y]) = 0,
| Ls | −F [X, Y], whichever is larger, | ls | = | ls | value of the color from the compressed image buffer 33,
The amount of compression in the color mapping table with the same H, Y index as F [X, Y] = | ls |, | ls ′ | = the color | ls after full color range compression has been performed by the color mapping device 34. | Value.

In order to limit the amount of saturation that occurs to a certain value (for example, the value M), the above equations (11) and (12) are used.
Instead of, the following equations (13) and (14) can be used. | Le '| = max (0, | le | -min (F [H, Y], M)) (13) | ls' | = max (0, | ls | -min (F [H, Y], M)) (14) where max (0, | le | -min (F [H,
Y], M)) = 0, | le | -min (F [H, Y],
The larger of M), max (0, | ls | -min (F
[H, Y], M)) = 0, | ls | -min (F [H,
Y], M) whichever is larger, min (F [H, Y], M) =
F [H, Y], the smaller of M, | le '|, | le |,
| Ls ′ |, | ls |, and F [H, Y] are expressed by equation (11),
It is the same variable as defined in (12).

Thus, the color mapping device 34 can control the compression of the entire source image color range so that any source image color that is desaturated beyond the extent determined by the M (eg, 50) value. Disappear. The value M can be changed according to the H and Y indexes, and can be constant for each compression amount stored in the color mapping table of the color mapping buffer 41. Formula (7),
Like the constants A, B, and N in (8), the value M is determined based on the input / output range of the equation (one or more) using this.

By varying the amount of desaturation applied to the source image pixel color as | le |, | ls |, and F [H, Y] values, the color mapping device 34 causes the source image to be less saturated. You can compress colors that are highly saturated (unlike colors).
In other words, the color mapping device 34 is small | le
Greater than the source image color with |, | ls | values |
Source image colors with le |, | ls | values can be compressed significantly.

The desaturation transformation described above can be performed by the color mapping device 34 using a desaturation function defined in a desaturation table (eg, a desaturation table stored in the color mapping table 41, not shown)). Each value in the desaturation table is indexed by the sum of (i) the corresponding value of the compression amount F [H, Y] and (ii) the maximum value of | le | and | ls | values. Using the desaturation table, each compression equation (11), (1
2) is replaced by the following equations (15) and (16). | Le '| = max (0, | le | -D {F [H, Y] + max (| le |, | ls |)}) (15) | ls' | = max (0, | ls | -D {F [H, Y] + max (| le |, | ls |)}) (16) where max (0, | le | −D {F [H, Y] + m
ax (| le |, | ls |)}) = 0, | le | -D
The larger of {F [H, Y] + max (| le |, | ls |)}), max (0, | ls | -D {F [H, Y] +
max (| le |, | ls |)}) = 0, | ls | -D
The larger one of {F [H, Y] + max (| le |, | ls |)}), D {F [H, Y] + max (| le |, | l
s |)}) = F [H, Y], max (| le |, | ls
Unsaturated values from the unsaturated table indexed by |), | le ′ |, | ls ′ |, | le |, | ls |, F
[H, Y] are the same variables as defined in equations (11) and (12).

By using the equations (15) and (16), the color mapping device 34 can change the desaturation amount based on the degree of saturation of each color of the source image. Thus, the further away the source image color coordinates are from the Y axis (ie, the more saturated the source image colors are), and the further the source image color coordinates are from the boundaries of the image output device full color gamut. The farther apart, the | le | and | ls | values of the color coordinates will be reduced or compressed by the color mapping device 34.

After the various colors of the source image are mapped or compressed by the color mapping device 34, the compressed values are output by the color mapping device 34 for printing (display if the image output device 35 is a CRT, etc.). It is output to the device 35. As mentioned above, the source image is read from the image buffer 33 while the compression performed by the color mapping device 34 is performed.

In the preferred embodiment, the full color gamut comparison device 37 assigns F [H, Y] values to the color mapping table 41 in the color mapping buffer 41 only for the H, Y indexes in the source image full color gamut table. To build the color mapping table. In order to compress a source image color that does not have the same index as one of the F [H, Y] values in the color mapping table from the image buffer 33, the color mapping device 34 performs an interpolation calculation to calculate the source image color. Determine the appropriate F [H, Y] values to subtract from the | le |, | ls | values. The interpolation calculation performed for a given source image color is F [H, H in the color mapping table indexed by the H, Y index corresponding to the color of the source image adjacent to that source image color from the image buffer 33. Performed by the color mapping device 34 using the Y] value. The advantage of this method is that the color mapping table stored in the color mapping buffer 41 does not require one F [H, Y] value for each color of the source image due to the interpolation calculation function.

In yet another embodiment, the full color gamut comparison device 37 provides F [H, Y] values for each H, Y index of the source image full color gamut table in the color mapping table in the color mapping buffer 41. The color mapping table is constructed by storing in 41. In other words, unlike the embodiment described above, the color mapping table contains all the F [H, Y] values needed to compress the source image stored in the image buffer 33, which results in an interpolation calculation by the color mapping device 34. The remaining F [H, Y] values are determined without the need to perform The color mapping table is F for each index of the source image full color range table.
Since the [H, Y] values are included, the source image is read from the image buffer 33, and at the same time, the color mapping device 34 does not perform the interpolation calculation.
| Le | for each color of the source image stored in
The F [H, Y] value can be subtracted from the | ls | value.

In any of the color mapping embodiments described above, the color inspection device 32 can store the colors of the source image in YES coordinates, Y, le, ls coordinates, absolute values of these coordinates, or in other formats. It should be noted that maintaining the source image colors in Y, le, ls format in the image buffer 33 is particularly useful in providing full color gamut compression decisions using embodiments that do not require interpolation calculations.

The above description relates to a method in which the entire color range of the source image can be compressed in the YES coordinate system using cylindrical geometry. However, the full color gamut compression performed by the system shown in FIG. 3 is also Y with spherical coordinates.
It can be performed in the ES coordinate system. Next, an example of compressing the entire color range of the source image in the YES coordinate system using spherical geometry rather than cylindrical geometry will be described.

As in the embodiment of the cylindrical geometric compression described above, at the same time when the source image is stored in the image buffer 33 (and the boundary of the source image full color range is stored in the source image full color range buffer 36). At the same time), the color space coordinates output by the image unit generator 31 are
Unless it is already in the YES coordinate format, it is converted into the YES coordinate and inspected by the color inspection unit 32.

In the inspection of the source image, the color inspection section 32a of the color inspection device 32 uses the equations (7) and (8).
Is used to determine the logarithmic value (for example, the base is 10) of the absolute value of the E and S coordinates for each color of the source image. As in the cylindrical geometry embodiment, a given color coordinate can be used to measure the degree of saturation of the color represented by that color coordinate.

Referring to FIG. 7, the compression line associated with a given color coordinate (eg point A) represented using cylindrical geometry is always horizontal and the Y value of the color coordinate (eg luminescence). ) From the point on the Y-axis to the color coordinates. However, as can be seen in FIG. 8, the compression line associated with a given color coordinate (eg, point A) represented using spherical geometry is not necessarily horizontal and is related to the Y value (eg, luminescence) of the color coordinate. Not extends from a single point on the Y axis to the color coordinates. Thus, full color gamut compression performed using spherical geometry in the YES coordinate system, although in some cases, is not necessarily performed along the horizon as in the cylindrical geometry embodiment.

In the spherical geometry embodiment, | le |, |
After determining the ls | value, the log value of the absolute value of (Y-C) is determined by the color inspection device 32 according to the following equation: ly = sign (Y−C) (A log (max (| Y−C |, N)) + B) (17) where ly = logarithmic value of absolute value of Y coordinate subtracted by constant C, sign (Y −C) = Sign of Y coordinate (+ or −) subtracted by constant C, max (| Y−C
|, N) = | Y−C |, the larger of N, C is a constant for subtracting Y (for example, 0.5), and A, B, and N are constants representing the Y value range and the ly value range. is there.

Like the equations (7) and (8), the constants A and B
Is a scaling constant (eg 60), an offset constant (eg 2.408) for expanding the ly range.
2). Further, like the equations (7) and (8), the constant N is a non-zero cutoff value (prevention of a decision error) when the (YC) value passes through 0 (that is, is equal to 0). For example, 0.0078125). Constants A, B, N
Is determined based on the input / output range of Expression (17). For example, when the range of the input Y is -0.5 to +0.5 and the desired range of the value of the output ly is -128 to +127,
Each constant A, B, N is 2.4082, 2.1, 0.00
390625. Preferably, the color inspection device 32 uses a look-up table to perform the Y value to ly conversion.

Instead of the absolute value of Y, the logarithmic value of the absolute value of (Y-C) is determined to bring the center of the compression sphere to a specific gray level. In particular, as shown in FIG. 8, using the logarithmic value of | Y−C | instead of | Y |
Specify according to the value of. The range of Y is 0-256,
If C is selected as 128, the center of the compression sphere is 50%
It becomes a gray level. Similarly, the range of Y is 0 to
When 1, the constant C is selected as 0.5, and the center of the compression sphere becomes a 50% gray level. Also, for other points on the Y axis other than the 50% gray level, by selecting different values for the constant C,
It can be used as the center of a compression sphere.

Similar to the cylindrical geometry embodiment, saturation-related values of the source image full color gamut table stored in the source image full color gamut buffer 36, determined for each color using equation (9) (eg, , | Le | value). However, instead of using the Y value as the other part of the index, the "W value" is determined and used. The W value can be used to measure the azimuth associated with a compression line from a point on the Y axis to a given color coordinate representing the source image color, and is determined by the color inspection device 32 according to the following equation:

W = max (| le |, | ls |)-| ly | (18) where max (| le |, | ls |)-| ly | = |
a value obtained by subtracting the factor of | ly | from the larger of le |, | ls |, | le | = the absolute value of the le coordinate with the source image color, | ls | = the absolute value of the ls coordinate with the source image color, | Ly | = the absolute value of the ly coordinate associated with the source image color.

In order to determine which quadrant the color coordinates of the source image color belong to based on the sign (E) value and the sign (S) value, the color checking device 32 uses the sign (Y-
C) value is used to determine the color coordinate by the value of a constant C on the Y axis of the YES coordinate system (for example, 50%
It is determined whether it is above or below according to the gray level. The color inspection device 32 uses the above information for each source image color to provide the source image full color range buffer 36.
Generate the source image full color gamut table at. The source image full color gamut table is a list of saturation-related values (eg, | le | values) for each color in the source image indexed by the value H alone or the values H and W.

Saturation related values stored in the source image full color range table can take the form of | le |, | ls |, | ly | values. In the preferred embodiment, the source image full color range table of source image full color range buffer 36 stores a maximum | le | value for each H, W index. In yet another embodiment, the source image full color range table of source image full color range buffer 36 stores a maximum | ls | value for each H, W index. In another embodiment, the source image full color range table of the source image full color range buffer 36 uses this value as | le | and | ls for each H and W index.
The maximum value of | and | ly | values is stored.

The source image full color gamut table generated by color inspection device 32 is stored in source image full color range buffer 36 by color inspection device 32. Similar to the cylindrical geometric compression embodiment, the full color gamut comparison device 37 compares the source image full color gamut table to a similarly indexed printer full color gamut stored in the image output device full color gamut buffer 40. The comparison result is used by the full color range comparison device 37 to construct a color mapping table including a list of full color range compression rates F [H, W] in the color mapping buffer 41. Some examples of color mapping tables constructed using spherical geometry are shown in Table 3.

[0095]

[Table 3] At the same time the source image is read by the color mapping device 34 from the image buffer 33 to the image output device 35 (ie, the first pixel of the source image is in the image buffer 3).
3 and before the last source image pixel has been received by the image output device 35), the color mapping device 34 (source image full color gamut table, saturated full color gamut table stored in the printer full color gamut table). The source image | le |, using the amount of compression in the color mapping table specified by the H, W indexes (such as values)
Decrease the | ls | and | ly | values. In particular, as the source image passes through the color mapping device 34, the color mapping device 34 causes the source image | le |, | ls |, |
The full color range compression amount is subtracted from the ly | value.

When the spherical geometric compression method is executed, it is preferable to store the source image in the image buffer 33 in the le, ls, ly coordinate format or | le |, | ls |, | ly | coordinate format. In this way, the compression amount stored in the color mapping table of the color mapping device 41 can be directly applied to the source image for the purpose of reducing the saturation value of the source image.

It should be noted that all of the above-mentioned modifications related to the cylindrical geometric compression method apply to the spherical geometric compression method. For example,
The desaturation table stored in the color mapping buffer 41 can be used in the spherical coordinate compression method to vary the desaturation amount applied to the source image pixel color based on the degree to which each pixel color of the source image is saturated. it can. The foregoing description of the preferred embodiment of the present invention has been made for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed, and variations and modifications can be made or obtained from the practice of the invention in light of the above teachings. The embodiments described above have been chosen and described to illustrate the principles of the invention and its practical application, but with various modifications to the invention in various embodiments suitable for the particular use contemplated by those skilled in the art. Available.
The scope of the present invention is defined by the claims and their equivalents.

[0098]

As described above, according to the present invention, the compression of the entire color range can be performed on the source image at the same time when the source image is read from the memory to the image output device such as a printer.

[Brief description of drawings]

FIG. 1 is a block diagram of a conventional color image printer system.

FIG. 2 is a block diagram of another conventional color image printer system.

FIG. 3 is a block diagram of a color image copying system including an image processing apparatus according to an embodiment of the present invention.

FIG. 4 is a flowchart showing a method executed by the image processing apparatus shown in FIG.

FIG. 5 is a diagram showing an outline of the entire color range of the source image and the entire color range of the printer included in the YES color coordinate system.

6 is a diagram along the Y-axis of the YES coordinate system shown in FIG.

FIG. 7 is a diagram showing a color space represented by a YES coordinate system using cylindrical geometry.

FIG. 8 is a diagram showing a color space represented by a YES coordinate system using spherical geometry.

[Explanation of symbols]

 11 ... Image input device 13 ... Image buffer 15 ... Image output device 17 ... Full color range compression circuit 21 ... Image unit generator 23 ... Image buffer 25 ... Image output device 31 ... Image unit generator 32 ... Color inspection device 32a, 32b ... Color inspection unit 33 ... Image buffer 34 ... Color mapping device 35 ... Image output device 36 ... Source image full color range 37 ... Full color range comparison device

Claims (1)

[Claims] 1. An image processing method performed by a data processing system having a memory, the first all-out output to be output by an image output device capable of outputting a second full color gamut of colors. Storing a multi-pixel image having a color in a color range in a memory of the data processing system; reading the multi-pixel image from the memory of the data processing system for reception by the image output device; The step of converting the colors of the multi-pixel image that are not in the second full color gamut to the colors of the second full color gamut after the start of and before completion of the reading step.